Using lentiviral vectors, we have established a novel method of inducing tumors by directly transducing the known genetic changes in appropriate cells. In the case of gliomas we have transduced as few as 50-60 astrocytes in the various regions of the brain of an adult immune competent mouse to form tumors. The transduced glial cells appear to undergo reprogramming to generate GBMs capable of self-renewal and differentiation to various cell types of the central nervous system. Furthermore, 15-40% of the gliomas can transdifferentiate to form functional endothelial cells lining the blood vessels in the deep part of the brain tumor. Based on these observations, we propose to pursue the following experiments in the next five years: Aim 1: Recapitulation of human GBM in a mouse model: In the original mouse model for glioma, as a proof of principle, we used H-Ras and myristoylated AKT as surrogates for EGFR activating and PTEN inactivating mutations. We now propose to make vectors with truncated EGFRviii mutant found in over 45% of human GBMs, PTEN siRNA (deletions found in 35% of GBMs) and p53 and RB pathways. Aim 2: Genomic analysis along tumor progression: To better characterize the glioma model we plan to perform a time- course genomic analysis to provide a detailed view of the genomic changes along the progression of the tumor, following 15 days, 4 weeks, 8 weeks or longer, post-injection of the recombinant virus. RNA extracted from the tumors will be analyzed by microarray analysis. Aim 3: Reprogramming of glial cells to neural stem cells: Transduction of glial cells in the hippocampus with H-Ras and sip53 lentiviral vector leads to the generation of Nestin+, neural stem cells (NSCs), which are capable of both self-renewal and differentiation to neurons, glia and oligodendrocytes. We propose to use some of the vectors generated in Aim 1 to reproduce this effect, because this appears to be a seminal event of reprogramming a committed differentiated cell to a stem cell, in this case a cancer initiating stem cell. Aim 4: Transdifferentiation of GBMs into endothelial cells: We have observed the existence of transdifferentiated endothelial cells (TDECs) in our lentiviral vector induced mouse GBM model. The TDECs express endothelial antigens CD31, CD34, and vWF in addition to tumor marker GFP. We want to confirm and extend these observations by confirmation of formation of TDECs in mouse GBMs generated by recombinant vectors containing various genetic changes observed in human tumors. We also plan to take human GBMs and see if they form GBMs when transplanted in mouse brain. We believe that the proposed experiments will allow us to understand the initiation, progression and maintenance of GBMs. We hope that our efforts to identify the targets will allow therapeutic intervention of this deadly disease.